[국내논문]녹두 유전자원 지방산 함량 대량평가를 위한 근적외선분광법의 적용 Determination of Seed Fatty Acids Using Near-Infrared Reflectance Spectroscopy(NIR) in Mung Bean(Vigna radiata) Germplasm원문보기
본 연구에서는 녹두 유전자원의 지방산 함량을 신속 대량 검정하는 기술을 개발하여 유전자원 활용 및 육종 촉진에 기여하고자 하였다. 유전자원 평가에 적합한 신속하고 비파괴적인 지방산 함량 평가기술을 개발하기 위해 공시자원 1,125점의 녹두 종자를 종실상태와 분쇄한 분말상태로 근적외선분광분석기(NIR)를 이용하여 1,104~2,494 nm에서의 스펙트럼을 얻고 이들 중 스펙트럼이 중복되지 않는 원산지가 다양한 대표자원 106점을 선발하여 일반적인 방법으로 지방산 함량을 분석하고, 이 값과 NIR 스펙트럼 흡광도값 간의 상관분석을 위한 calibration set로 활용하였다. 그 결과 palmitic acid, stearic acid, oleic acid, linoleic acid, linolenic acid 및 total fatty acid에 대한 NIR 흡광도와의 상관계수$R^2$이 각각 0.74, 0.18, 0.12, 0.72, 0.48 및 0.78로 나타났고, 이들 중 $R^2$가 높은 검량식을 미지의 시료 10점으로 검증한 결과, palmitic, linoleic 및 total fatty acid에 대한 검증 상관계수 $R^2$이 0.96, 0.74, 0.81로 나타나, 다양한 녹두 유전자원의 지방산함량 신속 대량 예측에 유효하게 활용될 수 있는 것으로 나타났다. 한편, 공시된 녹두 유전자원 115점 중에서 자원번호 IT208075 자원은 저 지방산 자원($14.24\;mg\;g^{-1}$)으로 선발되었고, IT163279 자원은 고 지방산 자원($18.43\;mg\;g^{-1}$)으로 선발되어 향후 녹두작물의 성분육종에 유용할 것으로 생각된다.
본 연구에서는 녹두 유전자원의 지방산 함량을 신속 대량 검정하는 기술을 개발하여 유전자원 활용 및 육종 촉진에 기여하고자 하였다. 유전자원 평가에 적합한 신속하고 비파괴적인 지방산 함량 평가기술을 개발하기 위해 공시자원 1,125점의 녹두 종자를 종실상태와 분쇄한 분말상태로 근적외선분광분석기(NIR)를 이용하여 1,104~2,494 nm에서의 스펙트럼을 얻고 이들 중 스펙트럼이 중복되지 않는 원산지가 다양한 대표자원 106점을 선발하여 일반적인 방법으로 지방산 함량을 분석하고, 이 값과 NIR 스펙트럼 흡광도값 간의 상관분석을 위한 calibration set로 활용하였다. 그 결과 palmitic acid, stearic acid, oleic acid, linoleic acid, linolenic acid 및 total fatty acid에 대한 NIR 흡광도와의 상관계수 $R^2$이 각각 0.74, 0.18, 0.12, 0.72, 0.48 및 0.78로 나타났고, 이들 중 $R^2$가 높은 검량식을 미지의 시료 10점으로 검증한 결과, palmitic, linoleic 및 total fatty acid에 대한 검증 상관계수 $R^2$이 0.96, 0.74, 0.81로 나타나, 다양한 녹두 유전자원의 지방산함량 신속 대량 예측에 유효하게 활용될 수 있는 것으로 나타났다. 한편, 공시된 녹두 유전자원 115점 중에서 자원번호 IT208075 자원은 저 지방산 자원($14.24\;mg\;g^{-1}$)으로 선발되었고, IT163279 자원은 고 지방산 자원($18.43\;mg\;g^{-1}$)으로 선발되어 향후 녹두작물의 성분육종에 유용할 것으로 생각된다.
* AI 자동 식별 결과로 적합하지 않은 문장이 있을 수 있으니, 이용에 유의하시기 바랍니다.
문제 정의
In this study, mung bean (Vigna radiata) germplasm collections were characterized and evaluated via NIR. Our objective was to investigate whether fatty acids (FAs) of a coarse cereal grain could be predicted by NIRs. Successful prediction will contribute to more effective application of NIRs in coarse grain germplasm use and crop breeding programs.
제안 방법
The reference values of all of the samples were used in mathematical treatments designed to create prediction equations using modified PLS. After using the equations to calculate predicted values, we conducted an external validation using an independent sample set (n=10) to ensure that the equations could be applied to the prediction of FAs. Fig.
1 compares the spectra of grain and flour foxtail millet germplasm. Because distinct small peaks were observed in raw spectra without mathematical treatment, these spectra were modified to examine the correlative wavelength and determine FAs via mathematical treatment of variable peaks and shapes. After this treatment, some absorption bands (e.
3. Histogram for the distribution of whole 1,125 accessions according to NIR predicted value of palmitic acid, linoleic acid and total FA contents by the equation developed in this study.
1,125 accessions of mung bean (Vigna radiata) germplasm conserved at the National Agrobiodiversity Center RDA genebank were NIR-scanned as grains. Next, the samples were ground, NIR-scanned again as flour, and then analyzed for FAs using gas chromatography (GC).
The data were centered on the mean spectrum and mean reference value using modified partial least-squares (PLS) regression. The PLS regression was modified so that reference values and reflectance data were scaled at each wavelength to a standard deviation of 1.0 before each PLS regression term. The preprocessing methods chosen for the model were the optimum to obtain minimum error following crossvalidation (four cross-validation groups per germplasm collection).
Preprocessing of the spectral data (1,104～2,494 ㎚) of 100 samples consisted of a normal multiplicative scatter. The data were centered on the mean spectrum and mean reference value using modified partial least-squares (PLS) regression. The PLS regression was modified so that reference values and reflectance data were scaled at each wavelength to a standard deviation of 1.
0 before each PLS regression term. The preprocessing methods chosen for the model were the optimum to obtain minimum error following crossvalidation (four cross-validation groups per germplasm collection). The optimum number of PLS regression terms for the calibration, also determined by cross-validation, was the number of factors yielding the minimum error between predicted and reference values (standard error of cross-validation, SECV).
The reference values of all of the samples were used in mathematical treatments designed to create prediction equations using modified PLS. After using the equations to calculate predicted values, we conducted an external validation using an independent sample set (n=10) to ensure that the equations could be applied to the prediction of FAs.
데이터처리
1, Downside). Comparison and investigation of the two spectral patterns was conducted using NIRs model regression. Zhang et al.
The modified PLS regression model was tested using independent validation samples (n=10). Statistics used to assess the model were the standard error of performance (SEP, not bias-corrected), coefficient of determination (R2), and slope and intercept of the linear regression of NIRs predicted versus analyzed values.
후속연구
The moisture condition of all of the germplasm conserved at the National Agrobiodiversity Center was assumed to be the same, as all germplasm materials at the center are conserved in identical mid-term conservation conditions of 4℃. A future study of the seed moisture conditions of each accession would be warranted.
참고문헌 (13)
Association of Official Analytical Chemists (AOAC). 1990. Official Methods of Analysis, 16th ed. Gaithersburg, MD
Cozzolino D, Kwiatkowski MJ, Parker M, Cynkar WU, Dambergs RG, Gishen M, Herderich MJ. 2004. Prediction of phenolic compounds in red wine fermentations by visible and near infrared spectroscopy. Analytica Chimica Acta 513:73-80
Gotor AA, Berger M, Farkas E, Labalette F, Centis S, Calmon A. 2007. Quantification of sunflower minor components by near infrared spectrometry(NIR). Helia 30:183-190
Jiang ZS, Ma C. 2008. Zhaoyuan of Shandong- vermicelli capital. China Food Quality News pp1
Kim KS, Park SH, Choung MG, Jang YS. 2007. Use of nearinfrared spectroscopy for estimating fatty acid composition in intact seeds of rapeseed. J Crop Sci Biotech 10:15-20
Kwak, CS, Lim SJ, Kim SA, Park SC, Lee MS. 2004. Antioxidative and antimutagenic effects of Korean buckwheat, sorghum, millet and job's tears. Korean Society of Food and Nutrition 7:921-1077
Kye IS, Jun YS, Cheigh HS. 1989. Functional properties of mung bean protein isolates. J Korean Soc Food Nutr 18:300-306
Lee SY, Lee YY, Lee SK, Cho YH, Ma KH, Kang HK, Gwag JG, Park KH, Lee HS. 2009. Selecting high amylose lines of rice varieties using NIR spectroscopy at the RDA gene bank conserved. Korean Journal of Breeding Science 41:108
Moschner C, Biskupek-Korel B. 2006. Estimating the content of free fatty acids in high oleic sunflower by near-infrared spectroscopy. Eur J Lipid Sci Technol 108:606-613
Rural Development Administration (RDA). 2009. The status of plant genetic resource conservation
United State Department of Agriculture. 2001. Nutrient database for standard reference. pp.14
Vines, L, Kays SE, Koehler PE. 2005. Near-infrared reflectance model for the rapid prediction of total fat in cereal foods. J Agric Food Chem 53:1550-1555
Zhang, C, Shen Y, Chen J, Xiao P and Bao J. 2008. Nondestructive prediction of total phenolics, flavonoid contents, and antioxidant capacity of rice grain using near-infrared spectroscopy. J Agric Food Chem 56:8268-8272
※ AI-Helper는 부적절한 답변을 할 수 있습니다.